Description

Given the focus on boosting retention rates and the potential benefits of pro-active and early identification of students who may require support, higher education institutions are looking at the data already captured in university systems to determine if they can be used to identify such students. This paper uses historical student data to validate an existing learning analytics tool, the Moodle Engagement Analytics Plugin (MEAP). We present data on the utility of the MEAP to identify students ‘at risk’ based on proxy measurements of online activity for three courses/units in three different disciplines. Our results suggest that there are real differences in the predictive power of the MEAP between different courses due to differences in the extent and structure of the learning activities captured in the learning management system.

Share

COinS
 

Validating the Effectiveness of the Moodle Engagement Analytics Plugin to Predict Student Academic Performance

Given the focus on boosting retention rates and the potential benefits of pro-active and early identification of students who may require support, higher education institutions are looking at the data already captured in university systems to determine if they can be used to identify such students. This paper uses historical student data to validate an existing learning analytics tool, the Moodle Engagement Analytics Plugin (MEAP). We present data on the utility of the MEAP to identify students ‘at risk’ based on proxy measurements of online activity for three courses/units in three different disciplines. Our results suggest that there are real differences in the predictive power of the MEAP between different courses due to differences in the extent and structure of the learning activities captured in the learning management system.